We introduce an algorithm for the deconvolution of radio synthesis images that accounts for the non-coplanar-baseline effect, allows multiscale reconstruction onto arbitrarily positioned pixel grids, and allows the antenna elements to have direcitonal dependent gains. Using numerical L1-minimisation techniques established in the application of compressive sensing to radio astronomy, we directly solve the deconvolution equation using GPU (graphics processing unit) hardware. This approach relies on an analytic expression for the contribution of a pixel in the image to the observed visibilities, and the well-known expression for Dirac delta function pixels is used along with two new approximations for Gaussian pixels, which allow for multi-scale deconvolution. The algorithm is similar to the CLEAN algorithm in that it fits the reconstructed pixels in the image to the observed visibilities while minimising the total flux; however, unlike CLEAN, it operates on the ungridded visibilities, enforces positivity, and has guaranteed global convergence. The pixels in the image can be arbitrarily distributed and arbitrary gains between each pixel and each antenna element can also be specified. Direct deconvolution of the observed visibilities is shown to be feasible for several deconvolution problems, including a 1 megapixel wide-field image with over 400,000 visibilities. Correctness of the algorithm is shown using synthetic data, and the algorithm shows good image reconstruction performance. Though this algorithm requires significantly more computation than methods based on the CLEAN algorithm, we demonstrate that it is trivially parallelisable across multiple GPUs and potentially can be scaled to GPU clusters. We also demonstrate that a significant speed up is possible through the use of multi-scale analysis using Gaussian pixels.

We have measured the muon flux at the underground Gran Sasso National Laboratory (3800 m w.e.) to be (3.41 \pm 0.01) \times 10-4m-2s-1 using four years of Borexino data. A modulation of this signal is observed with a period of (366\pm3) days and a relative amplitude of (1.29 \pm 0.07)%. The measured phase is (179 \pm 6) days, corresponding to a maximum on the 28th of June. Using the most complete atmospheric data models available, muon rate fluctuations are shown to be positively correlated with atmospheric temperature, with an effective coefficient {\alpha}T = 0.93 \pm 0.04. This result represents the most precise study of the muon flux modulation for this site and is in good agreement with expectations.

Borexino is a large-volume liquid scintillator detector installed in the underground halls of the Laboratori Nazionali del Gran Sasso in Italy. After several years of construction, data taking started in May 2007. The Borexino phase I ended after about three years of data taking. Borexino provided the first real time measurement of the $^{7}$Be solar neutrino interaction rate with accuracy better than 5% and confirmed the absence of its day-night asymmetry with 1.4% precision. This latter Borexino results alone rejects the LOW region of solar neutrino oscillation parameters at more than 8.5 $\sigma$ C.L. Combined with the other solar neutrino data, Borexino measurements isolate the MSW-LMA solution of neutrino oscillations without assuming CPT invariance in the neutrino sector. Borexino has also directly observed solar neutrinos in the 1.0-1.5 MeV energy range, leading to the first direct evidence of the $pep$ solar neutrino signal and the strongest constraint of the CNO solar neutrino flux up to date. Borexino provided the measurement of the solar $^{8}$B neutrino rate with 3 MeV energy threshold.

Borexino, a liquid scintillator detector at LNGS, is designed for the detection of neutrinos and antineutrinos from the Sun, supernovae, nuclear reactors, and the Earth. The feeble nature of these signals requires a strong suppression of backgrounds below a few MeV. Very low intrinsic radiogenic contamination of all detector components needs to be accompanied by the efficient identification of muons and of muon-induced backgrounds. Muons produce unstable nuclei by spallation processes along their trajectory through the detector whose decays can mimic the expected signals; for isotopes with half-lives longer than a few seconds, the dead time induced by a muon-related veto becomes unacceptably long, unless its application can be restricted to a sub-volume along the muon track. Consequently, not only the identification of muons with very high efficiency but also a precise reconstruction of their tracks is of primary importance for the physics program of the experiment. The Borexino inner detector is surrounded by an outer water-Cherenkov detector that plays a fundamental role in accomplishing this task. The detector design principles and their implementation are described. The strategies adopted to identify muons are reviewed and their efficiency is evaluated. The overall muon veto efficiency is found to be 99.992% or better. Ad-hoc track reconstruction algorithms developed are presented. Their performance is tested against muon events of known direction such as those from the CNGS neutrino beam, test tracks available from a dedicated External Muon Tracker and cosmic muons whose angular distribution reflects the local overburden profile. The achieved angular resolution is 3-5 deg and the lateral resolution is 35-50 cm, depending on the impact parameter of the crossing muon. The methods implemented to efficiently tag cosmogenic neutrons are also presented.

We present a new method of calculating the rate of neutrino-electron interactions in a strong magnetic field based on finite temperature field theory. Using this method, in which the effect of the magnetic field on the electron states is taken into account exactly, we calculate the rates of all of the lowest order neutrino-electron interactions in a plasma. As an example of the use of this technique, we explicitly calculate the rate at which neutrinos and antineutrinos annihilate in a highly magnetized plasma, and compare that to the rate in an unmagnetized plasma. The most important channel for energy deposition is the gyromagnetic absorption of a neutrino-antineutrino pair on an electron or positron in the plasma ($\nu\bar{\nu} e^\pm\leftrightarrow e^\pm$). Our results show that the rate of annihilation increases with the magnetic field strength once it reaches a certain critical value, which is dependent on the incident neutrino energies and the ambient temperature of the plasma. It is also shown that the annihilation rates are strongly dependent on the angle between the incident particles and the direction of the magnetic field. If sufficiently strong fields exist in the regions surrounding the core of a type II supernovae or in the central engines of gamma ray bursts, these processes will lead to more efficient plasma heating mechanism than in an unmagnetized medium, and moreover, one which is intrinsically anisotropic.

Finite temperature field theory is used to calculate the correction to the mass of the electron in plasma with finite temperature and arbitrary chemical potential, and the results are applied to the core regions of type II supernovae (SNe). It is shown that the effective electron mass varies between 1 MeV at the edge of the SN core up to 11 MeV near the center. This changed electron mass affects the rates of the electroweak processes which involve electrons and positrons. Due to the high electron chemical potential, the total emissivities and absorptivities of interactions involving electrons are only reduced a fraction of a percent. However, for interactions involving positrons, the emissivities and absorptivities are reduced by up to 20 percent. This is of particular significance for the reaction antineutrino + proton <-> positron + neutron which is a source of opacity for antineutrinos in the cores of type II SNe.

A standard version of a kinetic instability for the generation of Langmuir waves by a beam of electrons is adapted to describe the analogous instability due to a beam of neutrinos. The interaction between a Langmuir wave and a neutrino is treated in the one-loop approximation to lowest order in an expansion in $1/M_W^2$ in the standard electroweak model. It is shown that this kinetic instability is far too weak to occur in a suggested application to the reheating of the plasma behind a stalled shock in a type II supernova (SN). This theory is also used to test the validity of a previous analysis of a reactive neutrino beam instability and various shortcomings of this theory are noted. In particular, it is noted that relativistic plasma effects have a significant effect on the calculated growth rates, and that any theoretical description of neutrino-plasma interactions must be based directly on the electroweak theory. The basic scalings discussed in this paper suggest that a more complete investigation of neutrino-plasma processes should be undertaken to look for an efficient process capable of driving the stalled shock of a type II SN.